Variable Sparsity Kernel Learning

This paper(1) presents novel algorithms and applications for a particular class of mixed-norm regularization based Multiple Kernel Learning (MKL) formulations. The formulations assume that the given kernels are grouped and employ l(1) norm regularization for promoting sparsity within RKHS norms of each group and l(s), s >= 2 norm regularization for promoting non-sparse combinations across groups. Various sparsity levels in combining the kernels can be achieved by varying the grouping of kernels-hence we name the formulations as Variable Sparsity Kernel Learning (VSKL) formulations. While previous attempts have a non-convex formulation, here we present a convex formulation which admits efficient Mirror-Descent (MD) based solving techniques. The proposed MD based algorithm optimizes over product of simplices and has a computational complexity of O (m(2)n(tot) log n(max)/epsilon(2)) where m is no. training data points, n(max), n(tot) are the maximum no. kernels in any group, total no. kernels respectively and epsilon is the error in approximating the objective. A detailed proof of convergence of the algorithm is also presented. Experimental results show that the VSKL formulations are well-suited for multi-modal learning tasks like object categorization. Results also show that the MD based algorithm outperforms state-of-the-art MKL solvers in terms of computational efficiency.

Thermally Controlled Cyclic Insertion/Ejection of Dopant Ions and Reversible Zinc Blende/Wurtzite Phase Changes in ZnS Nanostructures

We report a reversible phase transformation of platelet-shaped ZnS nanostructures between wurtzite (WZ) and zinc blende (ZB) phases by reversible insertion/ ejection of dopant Mn(II) ions induced by a thermocyclic process. In a reaction flask loaded with WZ ZnS platelets and Mn molecular precursors, during heating Mn ions are incorporated and change the phase of the host nanostructures to ZB; during cooling Mn ions are spontaneously ejected, returning the host nanoplatelets to the original WZ phase. These reversible changes are monitored for several cycles with PL, EPR, XRD, and HRTEM. Interestingly, the (0001) WZ platelets transform to (110) ZB following a nucleation and growth process triggered by a local increase/depletion of the Mn2+ concentration in the nanocrystals.

Thermally Controlled Cyclic Insertion/Ejection of Dopant Ions and Reversible Zinc Blende/Wurtzite Phase Changes in ZnS Nanostructures

We report a reversible phase transformation of platelet-shaped ZnS nanostructures between wurtzite (WZ) and zinc blende (ZB) phases by reversible insertion/ ejection of dopant Mn(II) ions induced by a thermocyclic process. In a reaction flask loaded with WZ ZnS platelets and Mn molecular precursors, during heating Mn ions are incorporated and change the phase of the host nanostructures to ZB; during cooling Mn ions are spontaneously ejected, returning the host nanoplatelets to the original WZ phase. These reversible changes are monitored for several cycles with PL, EPR, XRD, and HRTEM. Interestingly, the (0001) WZ platelets transform to (110) ZB following a nucleation and growth process triggered by a local increase/depletion of the Mn2+ concentration in the nanocrystals.

The effect of surface mass transfer on buoyancy induced flow in a variable porosity medium adjacent to a vertical heated plate

The effect of surface mass transfer on buoyancy induced flow in a variable porosity medium adjacent to a heated vertical plate is studied for high Rayleigh numbers. Similarity solutions are obtained within the frame work of boundary layer theory for a power law variation in surface temperature,T Wpropx lambda and surface injectionv Wpropx(lambda–1/2). The analysis incorporates the expression connecting porosity and permeability and also the expression connecting porosity and effective thermal diffusivity. The influence of thermal dispersion on the flow and heat transfer characteristics are also analysed in detail. The results of the present analysis document the fact that variable porosity enhances heat transfer rate and the magnitude of velocity near the wall. The governing equations are solved using an implicit finite difference scheme for both the Darcy flow model and Forchheimer flow model, the latter analysis being confined to an isothermal surface and an impermeable vertical plate. The influence of the intertial terms in the Forchheimer model is to decrease the heat transfer and flow rates and the influence of thermal dispersion is to increase the heat transfer rate.

Gas-phase controlled mass transfer in a foam-bed reactor

Gas-phase controlled absorption of ammonia in foams made of solutions of sulphuric acid has been studied experimentally. Effects of gas-phase concentration of ammonia and type of surfactant on the performance of the foam-bed reactor are investigated. Gas-phase controlled absorption from a spherical bubble is anaylzed using the asymptotic value of Sherwood number (Sh = 6.58), for both negligible as well as significant changes in the volume of the bubble. The experimental data are shown to be in good agreement with the single-stage model of the foam-bed reactor using these asymptotic sub-models, as well as the diffusion-in-sphere analysis available in literature. Influence of effective diffusivity on the time dependence of fractional gas absorption has been found to be unimportant for foam columns with large times of contact. The asymptotic sub-models have been compared and use of the rigid-sphere asymptotic sub-model is recommended for foam columns of practical relevence.

Controlled synthesis of CuO nanostructures on Cu foil, rod and grid

CuO nanowires are synthesized by heating Cu foil, rod and grid in ambient without employing a catalyst or gas flow at temperatures ranging from 400 to 800 degrees C for a duration of 1-12 h. Scanning electron microscopy (SEM) investigation reveals the formation of nanowires. The structure, morphology and phase of the as-synthesized nanowires are analyzed by various techniques such as X-ray diffraction (XRD), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), thermogravimetric analysis (TGA) and Fourier transform infrared spectroscopy (FTIR). It is found that these nanowires are composed of CuO phase and the underlying film is of Cu2O. A systematic study is carried out to find the possibilities for the transformation of one phase to another completely. A possible growth mechanism for the nanowires is also discussed. (C) 2011 Elsevier B.V. All rights reserved.

Amorphous, Monoclinic, and Tetragonal Porous Zirconia Through a Controlled Self-Sustained Combustion Route

Porous, large surface area, metastable zirconias, are of importance to catalytic, electrochemical, biological, and thermal insulation applications. Combustion synthesis is a very commonly used method for producing such zirconias. However, its rapid nature makes control difficult. A simple modification has been made to traditional solution combustion synthesis to address this problem. It involves the addition of starch to yield a starting mixture with a ``dough-like'' consistency. Just 5 wt% starch is seen to significantly alter the combustion characteristics of the ``dough.'' In particular, it helps to achieve better control over reaction zone temperature that is significantly lower than the one calculated by the adiabatic approximation typically used in self-propagating high-temperature synthesis. The effect of such control is demonstrated by the ability to tune dough composition to yield zirconias with different phase compositions from the relatively elusive ``amorphous'' to monoclinic (> 30 nm grain size) and tetragonal pure zirconia (< 30 nm grain size). The nature of this amorphous phase has been investigated using infrared spectroscopy. Starch content also helps tailor porosity in the final product. Zirconias with an average pore size of about 50 mu m and specific surface area as large as 110 m2/g have been obtained.

Nonlinear dynamics and chaotic motions in feedback-controlled two- and three-degree-of-freedom robots

The dynamics of a feedback-controlled rigid robot is most commonly described by a set of nonlinear ordinary differential equations. In this paper we analyze these equations, representing the feedback-controlled motion of two- and three-degrees-of-freedom rigid robots with revolute (R) and prismatic (P) joints in the absence of compliance, friction, and potential energy, for the possibility of chaotic motions. We first study the unforced or inertial motions of the robots, and show that when the Gaussian or Riemannian curvature of the configuration space of a robot is negative, the robot equations can exhibit chaos. If the curvature is zero or positive, then the robot equations cannot exhibit chaos. We show that among the two-degrees-of-freedom robots, the PP and the PR robot have zero Gaussian curvature while the RP and RR robots have negative Gaussian curvatures. For the three-degrees-of-freedom robots, we analyze the two well-known RRP and RRR configurations of the Stanford arm and the PUMA manipulator respectively, and derive the conditions for negative curvature and possible chaotic motions. The criteria of negative curvature cannot be used for the forced or feedback-controlled motions. For the forced motion, we resort to the well-known numerical techniques and compute chaos maps, Poincare maps, and bifurcation diagrams. Numerical results are presented for the two-degrees-of-freedom RP and RR robots, and we show that these robot equations can exhibit chaos for low controller gains and for large underestimated models. From the bifurcation diagrams, the route to chaos appears to be through period doubling.

Variable temperature x-ray crystal structure analysis of a type I langbeinite: Rb2Cd2(SO4)(3)

The structure of a type I langbeinite, Rb2Cd2(SO4)(3), displays three different phases, cubic with a = 10.378(5) Angstrom (space group P2(1)3) at room temperature, monoclinic at 120 K with a = 10.328(3), b = 10.322(3), c = 10.325(3) Angstrom, beta = 89.975(1)degrees (space group P2(1)), and orthorhombic at 85 K with a = 10.319(2), b = 10.321(2), c = 10.320(2) Angstrom (space group P2(1)2(1)2(1)), respectively. Precise single-crystal analyses of these phases indicate that Rb2Cd2(SO4)(3) distorts initially from cubic to monoclinic upon cooling followed by a significant reorientation of the SO4 tetrahedra, resulting in an orthorhombic symmetry upon further cooling. The three structures have been established unequivocally using the same crystal. There is no indication of the formation of an intermediate triclinic phase or any lattice disorder as conjectured in several earlier reports on compounds belonging to the type I langbeinite. The bond valence sum analyses of the coordination around the Rb sites indicate asymmetry in the bond strengths which could be the driving force of the ferroelectric behavior in these materials.

A method of tracking the peak power points for a variable speed wind energy conversion system

In this paper, a method of tracking the peak power in a wind energy conversion system (WECS) is proposed, which is independent of the turbine parameters and air density. The algorithm searches for the peak power by varying the speed in the desired direction. The generator is operated in the speed control mode with the speed reference being dynamically modified in accordance with the magnitude and direction of change of active power. The peak power points in the P-omega curve correspond to dP/domega = 0. This fact is made use of in the optimum point search algorithm. The generator considered is a wound rotor induction machine whose stator is connected directly to the grid and the rotor is fed through back-to-back pulse-width-modulation (PWM) converters. Stator flux-oriented vector control is applied to control the active and reactive current loops independently. The turbine characteristics are generated by a dc motor fed from a commercial dc drive. All of the control loops are executed by a single-chip digital signal processor (DSP) controller TMS320F240. Experimental results show that the performance of the control algorithm compares well with the conventional torque control method.

Two-channel Kondo physics in odd impurity chains

We study odd-membered chains of spin-1/2 impurities, with each end connected to its own metallic lead. For antiferromagnetic exchange coupling, universal two-channel Kondo (2CK) physics is shown to arise at low energies. Two overscreening mechanisms are found to occur depending on coupling strength, with distinct signatures in physical properties. For strong interimpurity coupling, a residual chain spin-1/2 moment experiences a renormalized effective coupling to the leads, while in the weak-coupling regime, Kondo coupling is mediated via incipient single-channel Kondo singlet formation. We also investigate models in which the leads are tunnel-coupled to the impurity chain, permitting variable dot filling under applied gate voltages. Effective low-energy models for each regime of filling are derived, and for even fillings where the chain ground state is a spin singlet, an orbital 2CK effect is found to be operative. Provided mirror symmetry is preserved, 2CK physics is shown to be wholly robust to variable dot filling; in particular, the single-particle spectrum at the Fermi level, and hence the low-temperature zero-bias conductance, is always pinned to half-unitarity. We derive a Friedel-Luttinger sum rule and from it show that, in contrast to a Fermi liquid, the Luttinger integral is nonzero and determined solely by the ``excess'' dot charge as controlled by gate voltage. The relevance of the work to real quantum dot devices, where interlead charge-transfer processes fatal to 2CK physics are present, is also discussed. Physical arguments and numerical renormalization-group techniques are used to obtain a detailed understanding of these problems.

Defining the ‘modified Griffin plot’ in vortex-induced vibration: revealing the effect of Reynolds number using controlled damping

In the present work, we study the transverse vortex-induced vibrations of an elastically mounted rigid cylinder in a fluid flow. We employ a technique to accurately control the structural damping, enabling the system to take on both negative and positive damping. This permits a systematic study of the effects of system mass and damping on the peak vibration response. Previous experiments over the last 30 years indicate a large scatter in peak-amplitude data ($A^*$) versus the product of mass–damping ($\alpha$), in the so-called ‘Griffin plot’. A principal result in the present work is the discovery that the data collapse very well if one takes into account the effect of Reynolds number ($\mbox{\textit{Re}}$), as an extra parameter in a modified Griffin plot. Peak amplitudes corresponding to zero damping ($A^*_{{\alpha}{=}0}$), for a compilation of experiments over a wide range of $\mbox{\textit{Re}}\,{=}\,500-33000$, are very well represented by the functional form $A^*_{\alpha{=}0} \,{=}\, f(\mbox{\textit{Re}}) \,{=}\, \log(0.41\,\mbox{\textit{Re}}^{0.36}$). For a given $\mbox{\textit{Re}}$, the amplitude $A^*$ appears to be proportional to a function of mass–damping, $A^*\propto g(\alpha)$, which is a similar function over all $\mbox{\textit{Re}}$. A good best-fit for a wide range of mass–damping and Reynolds number is thus given by the following simple expression, where $A^*\,{=}\, g(\alpha)\,f(\mbox{\textit{Re}})$: \[ A^* \,{=}\,(1 - 1.12\,\alpha + 0.30\,\alpha^2)\,\log (0.41\,\mbox{\textit{Re}}^{0.36}). \] In essence, by using a renormalized parameter, which we define as the ‘modified amplitude’, $A^*_M\,{=}\,A^*/A^*_{\alpha{=}0}$, the previously scattered data collapse very well onto a single curve, $g(\alpha)$, on what we refer to as the ‘modified Griffin plot’. There has also been much debate over the last three decades concerning the validity of using the product of mass and damping (such as $\alpha$) in these problems. Our results indicate that the combined mass–damping parameter ($\alpha$) does indeed collapse peak-amplitude data well, at a given $\mbox{\textit{Re}}$, independent of the precise mass and damping values, for mass ratios down to $m^*\,{=}\,1$.

Synthesis and characterization of novel cationic lipid and cholesterol-coated gold nanoparticles and their interactions with dipalmitoylphosphatidylcholine membranes

Novel gold nanoparticles bearing cationic single-chain, double-chain, and cholesterol based amphiphilic units have been synthesized. These nanoparticles represent size-stable entities in which various cationic lipids have been immobilized through their thiol group onto the gold nanoparticle core. The resulting colloids have been characterized by UV-vis, (1)H NMR, FT-IR spectroscopy, and transmission electron microscopy. The average size of the resultant nanoparticles could be controlled by the relative bulkiness of the capping agent. Thus, the average diameters of the nanoparticles formed from the cationic single-chain, double-chain, and cholesterol based thiolate-coated materials were 5.9,2.9, and 2.04 nm, respectively. We also examined the interaction of these cationic gold nanoparticles with vesicular membranes generated from dipalmitoylphosphatidylcholine (DPPC) lipid suspensions. Nanoparticle doped DPPC vesicular suspensions displayed a characteristic surface plasmon band in their UV-vis spectra. Inclusion of nanoparticles in vesicular suspensions led to increases in the aggregate diameters, as evidenced from dynamic light scattering. Differential scanning calorimetric examination indicated that incorporation of single-chain, double-chain, and cholesteryl-linked cationic nanoparticles exert variable effects on the DPPC melting transitions. While increased doping of single-chain nanoparticles in DPPC resulted in the phases that melt at higher temperatures, inclusion of an incremental amount of double-chain nanoparticles caused the lowering of the melting temperature of DPPC. On the other hand, the cationic cholesteryl nanoparticle interacted with DPPC in membranes in a manner somewhat analogous to that of cholesterol itself and caused broadening of the DPPC melting transition.